You may not be able to upgrade from 3.{4,3,2,1,etc} from linux, but OS patches are no problem. I have a Vx running PalmOS 3.5 which I just patched w/ pilot-xfer. It comes as a.zip ('the Winblows version') or a.sit ('the Mac version') which decompresses to a.prc. At least they made this part easy.
It seems to me rather than not installing NFS (as some have suggested to you), you might use ipchains to disable NFS access from all but your trusted systems. I.e. just drop the NFS request packets from systems that aren't yours. You should probably do this for any of the other ports that you have open, too.
General Question: Is it just me, or shouldn't a well secured distribution behave like this out of the box?
It would be nice to see some of the open/free software packages move to a literate programming environment like web (or better yet noweb ). The code is the documentation, or alternately, the documentation is the code with these systems. I.e. LaTeX + Code in one file. If you've never coded this way, you should definitely give it a try.
At best the frames are 4000x4000x4 (i.e. 32bpp) or 61 MB a piece. This would give them about the same spatial resolution as a 35mm negative, and a comparable color depth. More likely they are 2Kx2K or 3Kx3K.
LN2 is often made by using an expansion piston process on air. I.e. the air is put into a cylinder and a pisto is drawn out which expands and cools the air. The air condenses and is drawn off. Then the LOx and other trace liquids are allowed to boil off, leaving the LN2 behind which is stored in a double walled/vaccuum insulated tank called a dewar (which others have noted already).
Yeah sure. Of course compressing nitrogen costs energy. Not to mention it can be difficult to move around. OTOH, the dewar will take care of most of the condesation/freezing issues. You can even run the a/c on the air that comes out of the heat exchanger!
We have to assume that the environmental cost of compressing nitrogen is smaller than the cost of burning gasoline in the car directly for this to have any hope of working. BTW, compressing N2 is a pretty cheap process.
Don't be silly. The energy in boiling the nitrogen (i.e. to take it to a higher state) comes from the air. LN2 at say 150K or so will warm up to 293K or so (depending on ambient temperature) just by being exposed to the ambient warmth. This expanding gas can drive a quite efficient little piston engine.
...people who want to make negative comments about an Apogee game will just have to skirt around the actual name of the game and, instead, say things like: 'That new Apogee game really sucked,' or ' Boy the Apogee game that came out on 4/1/2020 was really slow.'
Kind of annoying, but we can still trash their products. (UCITA still violates the 1st amendment and should be repealed or not adopted in the appropriate states, but this may be an option for those who can't afford a lawsuit.)
Actually the escape velocity is proportional to sqrt(g) so v_moon/v_earth = sqrt(g_moon/g_earth) = sqrt(1/6) = 0.4 or about 2.8 mi/sec, all other things being equal (which they aren't). Escape velocity also depends on radius of the body and the intial height of the spacecraft above the center of body. Lots of stuff to keep track of, you do the rest....
It may be too late for anyone to see this, but these guys have been doing this for years in Texas, Florida, Iowa, Wyoming, and some other states. They started in Texas, IIRC, and had to move off shore when they pissed some locals off. The compile drivers' license information, automobile registration, felony records, sexually related crimes, medical exams and other publicly available data. It's pretty cheap, and anyone can get and account.
let's see, 1e6 x 1e6 x 4bytes, thats 4e12 bytes.... hmm...
obiviously we don't have to store the whole thing, but that's not really the point.
matrix-vector products, also called matvec's (important in iterative solves of systems of linear algebraic equations) can be done faster in parallel than serially and are RAM bandwidth bound (not just amount). Then, after you do the part local to the processor, communications latency and bandwidth provide the bounds on performance.
Since I've put together one of these things, I thought I'd comment.....
The majority of the problems we like to solve are not the "embarrisingly parallel" problems like rendering frames of animation or bits of a single scene in a 3D picture. CFD (Computational Fluid Dynamics) applications (along with most other engineering apps) is matrix algebra intensive. Imagine matrices too large to fit into the RAM of a single processor (1e6 x 1e6 of 32-bit fp valuese, say). We divide this matrix up into sub-matrices and put the parts on the various nodes of the cluster. When we want to do matrix-vector multiplication there has to be a fair amount of communication for that to happen.
The point is that we require two things, high memory bandwidth (good cache helps here, but not enough), and low network latency. With fast ethernet the costs (in time) of sending one byte of data between two machines is nearly the same as 1k byte.
We don't particularly care about the OS as long as it stays out of the way. If the networking latency under BeOS is bad, then it's a bad OS for the cluster. If Be is constantly doing stuff in the backgroud, then it's a bad operating system for clusters.
There's nothing stopping anyone from porting MPI and giving it a try though.........
Slashdot Mirror
I would have sent this via email, but (no-email).....
Not to be to snide, but your employer does heavy-duty CFD in MATLAB?!?!?!?!. I hope these scripts are calling C or Fortran somewhere.
Seriously though, what kind of performance are you (they) getting out of MATLAB?
You may not be able to upgrade from 3.{4,3,2,1,etc} from linux, but OS patches are no problem. I have a Vx running PalmOS 3.5 which I just patched w/ pilot-xfer. It comes as a .zip ('the Winblows version') or a .sit ('the Mac version') which decompresses to a .prc. At least they made this part easy.
--
It seems to me rather than not installing NFS (as some have suggested to you), you might use ipchains to disable NFS access from all but your trusted systems. I.e. just drop the NFS request packets from systems that aren't yours. You should probably do this for any of the other ports that you have open, too.
General Question: Is it just me, or shouldn't a well secured distribution behave like this out of the box?
--
At best the frames are 4000x4000x4 (i.e. 32bpp) or 61 MB a piece. This would give them about the same spatial resolution as a 35mm negative, and a comparable color depth. More likely they are 2Kx2K or 3Kx3K.
For those who don't know,
LN2 is often made by using an expansion piston process on air. I.e. the air is put into a cylinder and a pisto is drawn out which expands and cools the air. The air condenses and is drawn off. Then the LOx and other trace liquids are allowed to boil off, leaving the LN2 behind which is stored in a double walled/vaccuum insulated tank called a dewar (which others have noted already).
Yeah sure. Of course compressing nitrogen costs energy. Not to mention it can be difficult to move around. OTOH, the dewar will take care of most of the condesation/freezing issues. You can even run the a/c on the air that comes out of the heat exchanger!
We have to assume that the environmental cost of compressing nitrogen is smaller than the cost of burning gasoline in the car directly for this to have any hope of working. BTW, compressing N2 is a pretty cheap process.
Don't be silly. The energy in boiling the nitrogen (i.e. to take it to a higher state) comes from the air. LN2 at say 150K or so will warm up to 293K or so (depending on ambient temperature) just by being exposed to the ambient warmth. This expanding gas can drive a quite efficient little piston engine.
There's also an emerge option in Emacs with a whole list of options (emerge-merge-directories bein, perhaps, the most helpful here).
...people who want to make negative comments about an Apogee game will just have to skirt around the actual name of the game and, instead, say things like: 'That new Apogee game really sucked,' or ' Boy the Apogee game that came out on 4/1/2020 was really slow.'
Kind of annoying, but we can still trash their products. (UCITA still violates the 1st amendment and should be repealed or not adopted in the appropriate states, but this may be an option for those who can't afford a lawsuit.)
Actually the escape velocity is proportional to sqrt(g) so v_moon/v_earth = sqrt(g_moon/g_earth) = sqrt(1/6) = 0.4 or about 2.8 mi/sec, all other things being equal (which they aren't). Escape velocity also depends on radius of the body and the intial height of the spacecraft above the center of body. Lots of stuff to keep track of, you do the rest....
Bill.
This will never be seen or moderated, but if Mr. Spielberg mowed lawns _and_ made great movies, I might be inclined to see more of them.
It may be too late for anyone to see this, but these guys have been doing this for years in Texas, Florida, Iowa, Wyoming, and some other states. They started in Texas, IIRC, and had to move off shore when they pissed some locals off. The compile drivers' license information, automobile registration, felony records, sexually related crimes, medical exams and other publicly available data. It's pretty cheap, and anyone can get and account.
let's see, 1e6 x 1e6 x 4bytes, thats 4e12 bytes.... hmm...
obiviously we don't have to store the whole thing, but that's not really the point.
matrix-vector products, also called matvec's (important in iterative solves of systems of linear algebraic equations) can be done faster in parallel than serially and are RAM bandwidth bound (not just amount). Then, after you do the part local to the processor, communications latency and bandwidth provide the bounds on performance.
Since I've put together one of these things, I thought I'd comment.....
The majority of the problems we like to solve are not the "embarrisingly parallel" problems like rendering frames of animation or bits of a single scene in a 3D picture. CFD (Computational Fluid Dynamics) applications (along with most other engineering apps) is matrix algebra intensive. Imagine matrices too large to fit into the RAM of a single processor (1e6 x 1e6 of 32-bit fp valuese, say). We divide this matrix up into sub-matrices and put the parts on the various nodes of the cluster. When we want to do matrix-vector multiplication there has to be a fair amount of communication for that to happen.
The point is that we require two things, high memory bandwidth (good cache helps here, but not enough), and low network latency. With fast ethernet the costs (in time) of sending one byte of data between two machines is nearly the same as 1k byte.
We don't particularly care about the OS as long as it stays out of the way. If the networking latency under BeOS is bad, then it's a bad OS for the cluster. If Be is constantly doing stuff in the backgroud, then it's a bad operating system for clusters.
There's nothing stopping anyone from porting MPI and giving it a try though.........